Comparison of methods by TLC and HPTLC for determination of aflatoxin M1 in milk and B1 in eggs

Milk and egg matrixes were assayed for aflatoxin M1 (AFM1) and B1 (AFB1) respectively, by AOAC official and modified methods with detection and quantification by thin layer chromatography (TLC) and high performance thin layer chromatography (HPTLC). The modified methods: Blanc followed by Romer, showed to be most appropriate for AFM1 analysis in milk. Both methods reduced emulsion formation, produced cleaner extracts, no streaking spots, precision and accuracy improved, especially when quantification was performed by HPTLC. The use of ternary mixture in the Blanc Method was advantageous as the solvent could extract AFM1 directly from the first stage (extraction), leaving other compounds in the binary mixture layer, avoiding emulsion formation, thus reducing toxin loss. The relative standard deviation (RSD%) values were low, 16 and 7% when TLC and HPTLC were used, with a mean recovery of 94 and 97%, respectively. As far as egg matrix and final extract are concerned, both methods evaluated for AFB1 need further studies. Although that matrix leads to emulsion with consequent loss of toxin, the Romer modified presented a reasonable clean extract (mean recovery of 92 and 96% for TLC and HPTLC, respectively). Most of the methods studied did not performed as expected mainly due to the matrixes high content of triglicerides (rich on saturated fatty acids), cholesterol, carotene and proteins. Although nowadays most methodology for AFM1 is based on HPLC, TLC determination (Blanc and Romer modified) for AFM1 and AFB1 is particularly recommended to those, inexperienced in food and feed mycotoxins analysis and especially who cannot afford to purchase sophisticated (HPLC,HPTLC) instrumentation.

Foresight of physical-chemical characteristics of apple juice blends appointed to sparkling drink elaboration

The purpose of this research was to combine the use of the component blend design to the response surface methodology, in order to foresee the effect of ternary apple juice blends (Catarina, Granny Smith and Pink Lady cultivars) on the physical-chemical characteristics of musts appointed to sparkling drink elaboration. Twelve mixes were made (three individual samples, three binary mixes and six ternary mixes), analyzed on the content of total reducing sugars, total titratable acidity and phenolic compounds; and adjusted, respectively, to the linear, quadratic and special cubic models. The results were organized in ternary charts of surface response and, from the overlap of these charts, it was determined a viable region which delimited the range of apple juice compositions that make musts physically and chemically suitable to sparkling drink elaboration. To represent the various possible combinations, the central point of the triangular area of the viable region was calculated and, this point, which represents the proportions of 23.22% of Catarina, 66.23% of Granny Smith and 10.55% of Pink Lady cultivars, was chosen to constitute the formulation of the must to be used in the elaboration of apple sparkling drinks.

Artificial neural networks (ANN): prediction of sensory measurements from instrumental data

The objective of this study was to predict by means of Artificial Neural Network (ANN), multilayer perceptrons, the texture attributes of light cheesecurds perceived by trained judges based on instrumental texture measurements. Inputs to the network were the instrumental texture measurements of light cheesecurd (imitative and fundamental parameters). Output variables were the sensory attributes consistency and spreadability. Nine light cheesecurd formulations composed of different combinations of fat and water were evaluated. The measurements obtained by the instrumental and sensory analyses of these formulations constituted the data set used for training and validation of the network. Network training was performed using a back-propagation algorithm. The network architecture selected was composed of 8-3-9-2 neurons in its layers, which quickly and accurately predicted the sensory texture attributes studied, showing a high correlation between the predicted and experimental values for the validation data set and excellent generalization ability, with a validation RMSE of 0.0506.

Um estudo sobre a vitimização para a cidade de São Paulo

A study about the victimization in the city of São Paulo. This paper applies the crime economics theory to Brazilian data. Following Becker (1968), Hinderlang et al. (1978) and Cohen et al. (1981), we tested the microeconomic factors that influence crime and victimization. For this end, the two waves of research of victimization of the Instituto Futuro Brasil, 2003 and 2008, were used in an effort to identify the determinants of victimization and police notification, using probit model. The main results suggest the factors which impact significantly the probability of victimization are the demographic characteristics, economic conditions and personal habits. The models of "life style" and "opportunity" seem to have good performance.